3250 cc. M5 KH phthalate +250 c.c. M/5 KCl +130 c.c. M/5 HCI +370 c.c. water.

6 20 c.c. M/5 KH phthalate +250 c.c. M/5 KCl +50 c.c. M/5 HCl +450 c.c. water.

$20 cc. Mỗ KII phthalate + 60 c.c. M/5 KOH+190 c.c. M/5 KCl +500 c.c. water. 20e.c. M.5 KH phthalate +150 c.c. M/5 KOH +100 c.c. M/5 KCI +500 c.c. water.

1 Socc. MKII phthalate +215 c.c. M/5 KOH + 35 c.c. M/5 KCl+500 c.c. water. 11 20 c.c. M/S KH phthalate +225 c.c. M/5 KOH+25 c.c. M/5 KCl +500 c.c. water.

12 230 c.c. M5 KH:PO, +20 c.c. M/5 KOH +230 c.c. M/5 KCl+500 c.c. water. 13 20 c.c. M5 KH;PO, +60 c.c. M/5 KOH+190 c.c. M/5 KCl +500 c.c. water. 14 20 c c. M5 KH;PO, +150 c.c. M/5 KOH+100 c.c. M/5 KCl+500 c.c. water. 15 C. M KHPO,+210 c.C. M/5 KOH +40 c.C. M/5 KC1+500 c.C. water. 250 e.c. M/5 KH;PO, +235 c.c. M/3 KOH +15 c.c. M/5 KCl+500 c.c. water.

16

18 125 c.c. M/5 HBO3+5 c.c. M/5 KOH +120 c.c. M/5 KCl +250 c.c. water. 19 125 c.c. M/5 HBO, +8 c.c. M/5 KOH +117 c.c. M/5 KCl +250 e.c. water. 20 125 c.c. M5 HzBO; +15 c.c. M/5 KOH +110 e.c. M/5 KCl +250 c.c. water. 125 c.e. M/5 H1BO; +40 c.c. M/5 KOH +85 c.c. M/5 KCl +250 c.c. water. 22 125 c.c. M5 H2BO; +80 c.c. M/5 KOH +45 c.c. M/5 KC1 +250 c.c. water. 23 125 c.c. M/5 H2BO2+120 c.c. M/5 KOH +5 c.c. M/5 KCl+20 c.c. water..

21

25 125 c.c. M/5 KOH +120 c.c. M/5 KH-PO1 +5 c.c. M/5 KCl +250 c.c. water. 26 125 c.c. M/S KOH +100 c.c. M/5 KH2PO, +25 c.c. M/5 KCl +250 c.c. water. T125 c.c. M/5 KOH +60 c.c. M/5 KH2PO4 +65 c.c. M/5 KCl +250 c.c. water. 28 125 c.c. M/5 KOH +0e.c. M/3 KH2PO1+125 c.c. M/5 KCl +250 c.c. water.

All measurements were made at 30° ±0.02. All potential differences recorded in this paper are converted to the normal hydrogen electrode standard on the assumption that the hydrogen electrode potential difference in M/20 KH phthalate is -0.2386. Hydrogen electrode measurements are designated by The Observed reduction

electrode potentials converted to the hydrogen standard are designated by E. The E value of a system of equal parts of total oxidant and of total reductant at constant pH is E'.. E', at pH=0 is designated by Eo, which, with the qualifications stated in our second paper, is the hypothetical normal potential of the system.

ELECTRODE MEASUREMENTS OF THE TETRASULPHONATE.

In Table III are given the results of a titration of reduced, buffered, indigo tetrasulphonate solution with dilute aqueous chromate solution.

TABLE III-Titration with chromate of 10 c. c. reduced 0.006 M indigo tetrasulphonate (No. 18) solution in 50 c. c. phthalate buffer No. 5. pil=2.817. Temp. 30°.

In is then 0.03006 log. En is the observed electrode potential difference reduced to the hydrogen electrode standard of reference.

The average value of E', is are given in the last column.

+0.1907, and the deviations from this These deviations are such as to indicate a fairly pure compound, the titration curve of which conforms closely to the ideal; but they are perhaps smaller than would be found if proper corrections were made for changes in pH during the course of the titration.

As already mentioned, corrections of potential for change in pH during titration were made in experiments with ferricyanide titration. An example of the consequences is shown in Table IV. A similar titration of sample 8 of the tetrasulphonate gave almost identical results. The data of Table IV are charted in Figure 2.

In Table V are given the results of a titration of the oxidant with a titanous solution. In this case the values of E', are not so consistent; but the titanium solution is difficult to adjust to known pH

and is not stable.

TABLE IV.-Reduced indigo tetrasulphonate (No. 18) titrated with K,FeCy. Five c. c. of approximately 0.006 M dye added to 50 c. c. buffer No. 7.

TABLE V.-Indigo tetrasulphonate (No. 8) titrated with Ti+++ solution. Both solution

buffered at pH 6.96.

TABLE VI.—Electrode potentials of mixtures of indigo tetrasulphonate (No. 18) and the reduction product, each added in 0.006 M solution to 50 c. c. buffer.

What is considered to be a better-controlled method for estimating E'. (the electrode potential of an equimolecular mixture of oxidant and reductant at constant pH) consists in adding oxidant and reductant in known proportions to a well-buffered solution, as described in the previous paper. In Table VI are given the results of a series of measurements with these solutions added, in the proportions indicated, to 50 c. c. of the buffer solutions. From the ratio of reductant to oxidant and the observed potential difference, En there is calculated the E', value for each case. This the hydrogen electrode potential of 50 c. c. buffer + 10 c. c. water. Assuming that this is the hypothetical hydrogen electrode potential of the dyebuffer mixture, we should expect, within the region of pH included, a constant difference E'-Th. Comment on the variation will be deferred to a later section of this paper.

Table VI shows the variation of E', with pH by means of only five instances. These serve to orient the system and the chart relating E', to pH may now be completed by using an approximately known mixture and carrying it through a wider range of pH.

3.

5.

TABLE VII-Electrode potentials of a mixture of indigo tetrasulphonate (No. 18) and its reduction product, at different pH values of the solution.

[10 c. c. 0.006 M mixture+50 c. c. buffer.]

9

13

14.

15.

20

22

25.

26.

27

In Table VII are shown the data for a mixture which proved to be 43.5 per cent reductant and 56.5 per cent oxidant. This fixed mixture was added in 10 c. c. quantities to 50 c. c. of buffer solution in each case. The pH and Th values given are those of the buffer solution+ 10 c. c. water. The En values observed are corrected to the corresponding E', values as follows: It is assumed that E'.for solution 1 is 0.3660, by Table VI. Consequently, in Table VII E' in the case of solution 1 is 0.2976. In this same case E is +0.3010. The difference, -0.0034, indicating that the mixture was 43.5 per cent reduced, was applied to En in all the other cases to give the values of E'..

In Figure 3, curve 4 shows the relation of E', to pH.

The smooth curve drawn through or near the plotted values was determined by means of an equation to be developed in a later section. The discrepancies between observed and calculated values will be discussed later.

ELECTRODE MEASUREMENTS OF THE TRISULPHONATE.

The first sample of what was supposed, from the method of preparation described in the literature, to be a trisulphonate, was found by titanium reduction to have the same properties as the tetrasulphonate and was so stated in our preliminary report (Sullivan and Clark, 1921). At that time the analyses were not complete. Further study indicated a mixture the effect of which, combined with the error of the titanium method, misled us. Accordingly, new preparations were made.

Preparation No. 23 gave the following results:

In Table VIII are given the results of a titration of the reduced solution with ferricyanide. The E values corrected for aciditychange are plotted against pH in Figure 2.